European countries, their dietary habits and diabetes prevalence

Introduction

Overview and Motivation

During the last 2 years, COVID-19 has been a main focus of the news. Though around 3% of the world population had COVID-19, diabetes can be considered as an even bigger health problem. Indeed, according to the International Diabetes Foundations (IDF), in 2019, 463 million adults were living with diabetes (around 6-7% of the world population) and this number is forecasted to rise to 700 million by 2050. Furthermore, 90% of cases of diabetes are of type 2, which means it results mainly from bad habits and not genetics. However both types of diabetes can be treated and/or prevented with a healthier diet and more physical activity. Additionally, according to the WHO, low income countries are more susceptible to having higher diabetes prevalence. Living in Europe, we observed that diabetes rates differ a lot from one country to another, so we wanted to find out if these rates were indeed linked to a country’s income, and if the nutritious composition of richer states’ population’s diet is also affected by this income difference and if yes, how it is affected.

Related Work

The article published by M. Adeva-Andany (2019) “Dietary habits contribute to define the risk of type 2 diabetes in humans” states that type 2 diabetes could be prevented by better dietary habits.They also found that higher animal protein consumption induced a greater risk of developing type 2 diabetes regardless of body mass index. In addition, they have also said that: “Low intake of animal products is linked to high educational level so that well-informed individuals tend to consume diets with elevated content of vegetable food” which can be linked to the GDP. (https://www.sciencedirect.com/science/article/pii/S240545771930419X?casa_token=Dzx3S9e8WgsAAAAA:IUYCy_uePTCOLUn38QUDO5KtpJopGpHdNIXZtHVVsGy9NWuXQ78Hp2juTvtjFQp67Dk2by-fMgPfrg)

Research questions

Therefore, we would like to find out answers to the following questions :

1. Do European countries that have higher GDPs really have lower diabetes prevalence ?

2. Do European countries that have higher GDPs consume less calories ?

3. How do the proportions of macronutrients (animal protein/plant protein/fat/carbohydrates) consumed differ between richer and poorer governments ?

4. And how do these differences relate to the diabetes prevalence in these countries ? What is the typical diet that can be observed in richer states that relates to lower diabetes prevalence ?

Data

To answer our research questions, we used three different datasets. While searching for datasets, we made sure that the years and countries matched for every one of them.

Wrangling and cleaning

Caloric consumption

The first dataset we used, downloaded from the portal https://ourworldindata.org/diet-compositions, contains information related to the supply of macronutrients in calories for different countries. We used data related to food supply rather than food consumption, as the latter is harder to find and generally, supply reflects the population’s demand and therefore its food consumption. The dataset gives us information on the average nutrition of different countries from 1961 to 2013 :

• It is composed of 8981 observations of 7 variables:

  • Entity Name of the country
    
  • Code ISO country code
    
  • YearYear of the observation
    
  • Calories from animal protein (FAO (2017)) The average per capita supply of calories derived from animal protein all measured in kilocalories per person per day
    
  • Calories from plant protein (FAO (2017)) The average per capita supply of calories derived from plant protein, all measured in kilocalories per person per day
    
  • Calories from fat (FAO (2017))The average per capita supply of calories derived from fat, all measured in kilocalories per person per day
    
  • Calories from carbohydrates (FAO (2017)) The average per capita supply of calories derived from carbohydrates, all measured in kilocalories per person per day

The intake of specific macronutrients (carbohydrates, protein and fats) are derived based on average food composition factors – these factors are derived and presented in the Food and Agriculture Organisation’s (FAO) Food Balance Sheet Handbook (https://www.fao.org/faostat/en/#data).

We will only focus on observations of European countries in the 2000s.

We used the ISO code as it is standardized worldwide and does not have the risk of having different names in different tables like the countries’ names.

Then, we proceeded to compute the mean of the consumption for each type of macronutrient in each country between the years 2000 and 2013, and the sum of total calories per person per day for each country in order to answer our second research question.

We then created a new table by adding the sum of total calories per person per day for each country in order to get a broader view with the total consumption of calories. To make sure that the joining of tables go smoothly, we also removed duplicates and the country name column.

Our assumption was that a county’s wealth may fluctuate over the course of 10 years (ex: a dip during the economic crisis of 2008) but an overall mean is sufficient to compare the different countries and their riches.

• We now have a dataframe with the following variables :

  • country_code ISO country code
    
  • cal_prot_animalThe mean of the calories from animal protein consumed per person in each country in the years 2000-2013
    
  • cal_prot_plant The mean of the calories from plant protein consumed per person in each country in the years 2000-2013
    
  • cal_carbsThe mean of the calories from carbohydrates consumed per person in each country in the years 2000-2013
    
  • cal_fat The mean of the calories from fat consumed per person in each country in the years 2000-2013
    
  • total_consumption The total calorie consumption per person based on the means of the consumption of each type of macronutrients in each countries in the years 2000-2013

The final table is the following:

Table 1: Caloric Consumption

Country Code	Calories from animal protein	Calories from plant protein	Calories from carbohydrates	Calories from fat	Total consumption
AUT	245	169	1833	1454	3702
BEL	238	158	1856	1467	3719
BGR	155	168	1606	846	2775
HRV	168	148	1691	940	2946
CYP	197	127	1291	1019	2633
CZE	218	153	1728	1155	3254
DNK	273	157	1746	1190	3366
EST	212	167	1967	842	3188
FIN	269	166	1623	1177	3234
FRA	293	161	1611	1480	3545
DEU	240	159	1785	1276	3460
GRC	250	204	1744	1338	3536
HUN	193	153	1560	1221	3126
IRL	279	174	1950	1187	3590
ITA	241	204	1776	1390	3612
LVA	203	154	1687	1044	3087
LTU	274	198	2055	858	3385
LUX	288	148	1752	1318	3507
MLT	237	204	1932	994	3367
NLD	292	136	1599	1195	3222
POL	204	197	1969	1035	3405
PRT	275	177	1824	1240	3516
ROU	200	220	2003	916	3340
SVK	140	150	1610	952	2853
SVN	230	168	1664	1067	3129
ESP	279	159	1481	1323	3243
SWE	285	143	1566	1137	3131
CHE	237	138	1660	1392	3426
GBR	232	178	1748	1256	3414

GDP

Our second dataset, downloaded from the portal https://data.worldbank.org, gives us information about the GDP of many countries over the course of 60 years (1960-2020).

• It is composed of 266 observations of 65 variables :

  • Country Name Name of the country
    
  • Country Code ISO country code
    
  • Indicator Name equal to “GDP in current US$” for every row
    
  • Indicator Code equal to “NY.GDP.MKTP.CD” for every row
    
  • And a variable for each year from 1960 to 2020

As we can see below, RStudio imported the Excel file as is, and so our column names found themselves at the 3rd row and therefore column names of columns 3 to 65 have been attributed numbers.

We decided to fix that and to filter out the years that is in our interest and that we have in common with other tables, so the years 2000-2013. We decided to get rid of the Indicator Name and Indicator Code variables as well since the values are the same for every row and they do not provide additional useful information.

Now, we want to filter out the European countries, just like in the first table :

In order to join tables easily, we transformed the columns corresponding to different years to a single “year” column, in order to have at each row of this dataset the GDP of a certain country at a certain year.

To make it easier to manipulate data, we decided to rename our variables for this table as well. We also made sure that the type of our numeric variable (GDP) was numeric and not character, like it was by default. In order to have graphs that are easy to read in the exploratory data analysis, we also decided to divide the avg_gdp column by a billion.

Lastly, we computed the average GDP for each country over the years 2000-2013 in order to be able to plot different variables together.

Table 2: GDP

Country Name	Country Code	Average GDP (in billion $)
Austria	AUT	335.98
Belgium	BEL	406.97
Bulgaria	BGR	37.41
Croatia	HRV	48.12
Cyprus	CYP	20.15
Czech Republic	CZE	158.48
Denmark	DNK	275.37
Estonia	EST	16.47
Finland	FIN	216.58
France	FRA	2283.48
Germany	DEU	3003.51
Greece	GRC	246.19
Hungary	HUN	110.92
Ireland	IRL	203.04
Italy	ITA	1872.80
Latvia	LVA	21.04
Lithuania	LTU	30.77
Luxembourg	LUX	42.85
Malta	MLT	7.27
Netherlands	NLD	721.70
Poland	POL	365.43
Portugal	PRT	200.37
Romania	ROU	125.08
Slovak Republic	SVK	70.82
Slovenia	SVN	39.50
Spain	ESP	1176.91
Sweden	SWE	425.83
Switzerland	CHE	490.18
United Kingdom	GBR	2416.76

We now have a dataframe with the following variables :

• country_name name of the country
  
• country_code ISO code of the country
  
• avg_gdp the average GDP of a country over the course of 2000-2013

Population

Since we will be observing the relation between the GDP with the calories consumed per person, it could be useful to have the GDP per person for the analysis. This is why we will be importing this dataset from https://data.worldbank.org/indicator/SP.POP.TOTL which gives us information on the evolution of the population per country over 1960-2020.

• It is composed of 266 observations of 65 variables :

  • Country Name Name of the country
    
  • Country Code ISO country code
    
  • Indicator Name equal to “Population, total” for every row
    
  • Indicator Code equal to “SP.POP.TOTL” for every row
    
  • And a variable for each year from 1960 to 2020

As this dataset comes from the same source and is the same file type as GDP, we can proceed with the same wrangling

In order to analyze the link between the GDP per person and calorie consumption per person, we will create a separate table which we will join to the final clean dataset.

We now have a dataframe with the following variables :

• country_name name of the country
  
• country_code ISO code of the country
  
• gdp_per_person the average GDP per person of a country over the course of 2000-2013

Diabetes

The third dataset, downloaded from https://www.ncdrisc.org/data-downloads-diabetes.html, gives us information about the age-standardised diabetes prevalence for each country and gender from 1980 to 2014.

• It is composed of 14’000 observations for 7 variables :

  • Country/Region/World Name of the country
    
  • ISO ISO country code
    
  • Sex Gender for which the diabetes prevalence is measured in a certain country at a certain year
    
  • Year Year of observation (1980-2014)
  • Age-standardised diabetes prevalence Diabetes rate considering all ages
    
  • Lower 95% uncertainty interval Lower confidence interval limit for the diabetes rate
  • Upper 95% uncertainty interval Higher confidence interval limit for the diabetes rate

Like with the first 2 datasets, we filtered our data to keep only European countries and observations between the years 2000 and 2013.

We also decided not to use the 95% confidence interval variable.

Then, we separated our dataset into two subsets. One with data about men.

Another one with data about women.

We then changed the variable names of these 2 subsets to facilitate joining tables later on.

Finally we grouped observations by country to get the mean prevalence/rate of diabetes between 2000 and 2013 for each European country :

• For men :

Table 3: Diabetes rates for men

Country Code	Diabetes rate
AUT	0.053
BEL	0.057
BGR	0.073
CHE	0.050
CYP	0.077
CZE	0.078
DEU	0.059
DNK	0.055
ESP	0.084
EST	0.071
FIN	0.066
FRA	0.071
GBR	0.063
GRC	0.069
HRV	0.071
HUN	0.080
IRL	0.069
ITA	0.065
LTU	0.078
LUX	0.068
LVA	0.071
MLT	0.088
NLD	0.052
POL	0.074
PRT	0.075
ROU	0.062
SVK	0.072
SVN	0.066
SWE	0.058

• For women :

Table 4: Diabetes rates for women

Country Code	Diabetes rate
AUT	0.053
BEL	0.057
BGR	0.073
CHE	0.050
CYP	0.077
CZE	0.078
DEU	0.059
DNK	0.055
ESP	0.084
EST	0.071
FIN	0.066
FRA	0.071
GBR	0.063
GRC	0.069
HRV	0.071
HUN	0.080
IRL	0.069
ITA	0.065
LTU	0.078
LUX	0.068
LVA	0.071
MLT	0.088
NLD	0.052
POL	0.074
PRT	0.075
ROU	0.062
SVK	0.072
SVN	0.066
SWE	0.058

We now have 2 dataframes with the following variables :

• country_code ISO code of the country
  
• prop_men_diabetes or prop_women_diabetesthe average diabetes rate in a country in the 2000-2013 timeframe

Joining tables

For the last step of our tidying, we joined all four tables in one dataset with the country_code key :

Table 5: Joint table for GDP, diabetes and caloric consumption

Country Name	Country Code	Average GDP (in billion $)	GDP per person (in $)	Men Diabetes	Women Diabetes	Calories from animal protein	Calories from plant protein	Calories from carbohydrates	Calories from fat	Total consumption
Austria	AUT	335.98	40545	0.053	0.034	245	169	1833	1454	3702
Belgium	BEL	406.97	38015	0.057	0.039	238	158	1856	1467	3719
Bulgaria	BGR	37.41	4988	0.073	0.064	155	168	1606	846	2775
Croatia	HRV	48.12	11189	0.071	0.059	168	148	1691	940	2946
Cyprus	CYP	20.15	18890	0.077	0.056	197	127	1291	1019	2633
Czech Republic	CZE	158.48	15283	0.078	0.065	218	153	1728	1155	3254
Denmark	DNK	275.37	50213	0.055	0.035	273	157	1746	1190	3366
Estonia	EST	16.47	12283	0.071	0.064	212	167	1967	842	3188
Finland	FIN	216.58	40806	0.066	0.044	269	166	1623	1177	3234
France	FRA	2283.48	35700	0.071	0.044	293	161	1611	1480	3545
Germany	DEU	3003.51	36733	0.059	0.040	240	159	1785	1276	3460
Greece	GRC	246.19	22344	0.069	0.060	250	204	1744	1338	3536
Hungary	HUN	110.92	11050	0.080	0.063	193	153	1560	1221	3126
Ireland	IRL	203.04	46899	0.069	0.049	279	174	1950	1187	3590
Italy	ITA	1872.80	32000	0.065	0.047	241	204	1776	1390	3612
Latvia	LVA	21.04	9774	0.071	0.065	203	154	1687	1044	3087
Lithuania	LTU	30.77	9697	0.078	0.069	274	198	2055	858	3385
Luxembourg	LUX	42.85	87546	0.068	0.039	288	148	1752	1318	3507
Malta	MLT	7.27	17759	0.088	0.066	237	204	1932	994	3367
Netherlands	NLD	721.70	43883	0.052	0.037	292	136	1599	1195	3222
Poland	POL	365.43	9586	0.074	0.066	204	197	1969	1035	3405
Portugal	PRT	200.37	19077	0.075	0.052	275	177	1824	1240	3516
Romania	ROU	125.08	6062	0.062	0.059	200	220	2003	916	3340
Slovak Republic	SVK	70.82	13146	0.072	0.059	140	150	1610	952	2853
Slovenia	SVN	39.50	19502	0.066	0.065	230	168	1664	1067	3129
Spain	ESP	1176.91	26263	0.084	0.059	279	159	1481	1323	3243
Sweden	SWE	425.83	46181	0.058	0.040	285	143	1566	1137	3131
Switzerland	CHE	490.18	64036	0.050	0.030	237	138	1660	1392	3426
United Kingdom	GBR	2416.76	39338	0.063	0.049	232	178	1748	1256	3414

Missing values

We did not have any NA values in our tables, we think this is due to the fact that we really spent time on gathering quality data that matched in terms of dates and countries.

Exploratory data analysis

First, even though we will be taking the means of the variables with which we are trying to answer our questions, it is interesting to observe their evolution in each country over time. We started with the GDP.

Evolution of GDP per country

We can see that the GDP of France, Germany, Italy, Spain and the United Kingdom had a significant increase between 2000 and 2008.

Plotting GDP against Diabetes (Men & Women)

Now let’s see if there is a relation between the GDP of a country and its diabetes prevalence. (men = blue, women = red)

We observe that apart of 5 outliers, our observations are mostly bunched up at the left of the graph. We decided to exclude these 5 observations, to see if we can observe a trend with the other countries. These outliers, as we can see on the graph before, are the countries that had a big increase of GDP in the time period of 2000-2013.

Without the outliers, we can see a bit more clearly. Indeed, it seems that the richer a country is, the lesser it has a high diabetes rate among its population.

Evolution of the consumption of macronutrients in calories per country

For the second table, we tried to see again if there was a trend in the consumption of different macro-nutrients in the 2000s for each country in our sample.

In different countries, there is one difference that stands out and that seems to be related to wealth. Indeed, countries with a higher GDP like Austria consume on average more fat as can be seen on this graph:

Whereas, countries with a lower GDP like Bulgaria have a lower fat consumption, as seen below:

There do not seem to be any trends in the graphs above and diets seem rather stable in each country, which is why we will take the average consumption for each macro-nutrient for our analysis. We can however note that the 5 outliers mentioned before tend to have a higher fat consumption than the countries with a smaller GDP.

Plotting GDP against calories consumed

We then wanted to analyse the relation between a country’s GDP and its individual consumption of each macronutrient as well as its total calorie consumption to see if there’s a trend.(total calories = orange, fat = blue, carbohydrates = purple, animal protein = red, plant protein = green)

We see that the calorie consumption does not really change. We wanted a close up on the relation between the total calorie consumption with the GDP for each country to see if we can spot outliers again, so we created other plots.

We end up again with these 5 outliers that have a higher than average GDP so if we remove them, we obtain the following plots :

Now we can more easily state that there’s a trend. It appears that the higher a country’s GDP, the higher the total calories consumed, contrary to our hypothesis.

Evolution of Diabetes per country

Once again, we tried to see if the diabetes prevalence in each country changed over the years 2000-2013.

We saw right away that the prevalence of diabetes is higher for man than women across all countries (there are however two exceptions : in Romania between 2000 and 2003 and Slovenia between 2000 and 2006).

We observed three different scenarios for countries that we selected: A decrease of diabetes over time for:

• Belgium
  
• Denmark
  
• Finland

We take Belgium as an example :

A decrease over time for women but not for men for :

• Austria
  
• Malta
  
• Netherlands
  
• Germany
  
• Italy
  
• Spain
  
• Switzerland

We take Austria as an example :

In other European countries, the prevalence of diabetes is increasing (at different paces) over time.

We take Croatia as an example :

Plotting Diabetes against each type of macronutrient consumption

Finally, we want to plot the relation between the diabetes prevalence against the total calorie consumption as well as each type of macronutrient consumed.

We can see a negative trend for the total consumption, the calories from animal protein and the calories from fat. We can observe a positive trend against calories from plant protein. For protein from carbohydrates, we can see a slighty positive trend for women.

Plotting Diabetes against each type of macronutrient consumption (without outliers)

Now, since they affected our plots that included the GDP variable so much, we want to see if we have different trends when we remove our 5 outliers.

Without our 5 outliers, we observe not much change in the trend of each type of calories consumed apart for carbohydrates where the trend changes for men and become slightly positive.

Analysis

1.Do European countries that have higher GDPs really have lower diabetes prevalence ?

This first question serves more as a control, since we learned during our research prior to our project that countries with higher GDPs tend to have lower diabetes rates. Indeed, we can observe that in the EDA.

It is important to note that, when we try to fit a linear model on these variables and observe correlations over all observations, we see that these relationships are not significant at all.

**Correlation with outliers

Between avg_gdp and women diabetes	-0.369
Between avg_gdp and men diabetes	-0.236

	Diabetes vs Women diabetes
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	1847.35	654.70	2.82	0.009
prop women diabetes	-25161.76	12203.46	-2.06	0.049
Observations	29
R2 / R2 adjusted	0.136 / 0.104

	Diabetes vs Men diabetes
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	1890.37	1086.48	1.74	0.093
prop men diabetes	-19988.04	15812.10	-1.26	0.217
Observations	29
R2 / R2 adjusted	0.056 / 0.021

However, once we exclude outliers, we see that the relationship is way more significant !

**Correlation without outliers

Between avg_gdp and women diabetes	-0.696
Between avg_gdp and men diabetes	-0.739

	Diabetes vs Women diabetes (without outliers
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	741.51	123.89	5.99	<0.001
prop women diabetes	-10299.65	2263.83	-4.55	<0.001
Observations	24
R2 / R2 adjusted	0.485 / 0.461

	Diabetes vs Men diabetes (without outliers
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	1147.62	187.48	6.12	<0.001
prop men diabetes	-14048.33	2730.12	-5.15	<0.001
Observations	24
R2 / R2 adjusted	0.546 / 0.526

In the EDA section, we grouped countries in 3 categories, according to the relationship between the GDP and diabetes rate. Here, we confirmed statistically that a relationship exists between these 2 variables when we remove outliers. This therefore made us think that these countries could be categorized into clusters.

To determine the number of clusters we used the elbow method. This method examines the percentage of variance explained as a function of the number of clusters. It is based on the idea that a number of clusters should be chosen such that the addition of another cluster does not allow for a better modeling of the data. The percentage of variance explained by the clusters is plotted against the number of clusters.

We therefore see from the graph above that the optimal number of clusters is 3. The allocation of countries according to their cluster is therefore as follows:

Table 6: Cluster 1

	avg_gdp	gdp_per_person	prop_men_diabetes	prop_women_diabetes	cal_prot_animal	cal_prot_plant	cal_carbs	cal_fat	total_consumption	cluster
AUT	336	40545	0.053	0.034	245	169	1833	1454	3702	1
BEL	407	38015	0.057	0.039	238	158	1856	1467	3719	1
DNK	275	50213	0.055	0.035	273	157	1746	1190	3366	1
FIN	217	40806	0.066	0.044	269	166	1623	1177	3234	1
FRA	2283	35700	0.071	0.044	293	161	1611	1480	3545	1
DEU	3004	36733	0.059	0.040	240	159	1785	1276	3460	1
IRL	203	46899	0.069	0.049	279	174	1950	1187	3590	1
ITA	1873	32000	0.065	0.047	241	204	1776	1390	3612	1
NLD	722	43883	0.052	0.037	292	136	1599	1195	3222	1
SWE	426	46181	0.058	0.040	285	143	1566	1137	3131	1
CHE	2417	39338	0.063	0.049	232	178	1748	1256	3414	1

Table 7: Cluster 2

	avg_gdp	gdp_per_person	prop_men_diabetes	prop_women_diabetes	cal_prot_animal	cal_prot_plant	cal_carbs	cal_fat	total_consumption	cluster
BGR	37.41	4988	0.073	0.064	155	168	1606	846	2775	2
HRV	48.12	11189	0.071	0.059	168	148	1691	940	2946	2
CYP	20.15	18890	0.077	0.056	197	127	1291	1019	2633	2
CZE	158.48	15283	0.078	0.065	218	153	1728	1155	3254	2
EST	16.47	12283	0.071	0.064	212	167	1967	842	3188	2
GRC	246.19	22344	0.069	0.060	250	204	1744	1338	3536	2
HUN	110.92	11050	0.080	0.063	193	153	1560	1221	3126	2
LVA	21.04	9774	0.071	0.065	203	154	1687	1044	3087	2
LTU	30.77	9697	0.078	0.069	274	198	2055	858	3385	2
MLT	7.27	17759	0.088	0.066	237	204	1932	994	3367	2
POL	365.43	9586	0.074	0.066	204	197	1969	1035	3405	2
PRT	200.37	19077	0.075	0.052	275	177	1824	1240	3516	2
ROU	125.08	6062	0.062	0.059	200	220	2003	916	3340	2
SVK	70.82	13146	0.072	0.059	140	150	1610	952	2853	2
SVN	39.50	19502	0.066	0.065	230	168	1664	1067	3129	2
ESP	1176.91	26263	0.084	0.059	279	159	1481	1323	3243	2

Table 8: Cluster 3

	avg_gdp	gdp_per_person	prop_men_diabetes	prop_women_diabetes	cal_prot_animal	cal_prot_plant	cal_carbs	cal_fat	total_consumption	cluster
LUX	42.9	87546	0.068	0.039	288	148	1752	1318	3507	3
GBR	490.2	64036	0.050	0.030	237	138	1660	1392	3426	3

It is important to note that our 3rd cluster only contains 2 countries, and therefore when we analyse correlation and linear regression parameters, we will be mainly considering the overall nature of the relationships. (positive or negative link?)

Now let’s plot these clusters to see the differences between them.

In the graph above,

• Group 1 = Cluster 3
• Group 2 = Cluster 2
• Group 3 = Cluster 1

We can therefore see that diabetes is indeed lower in clusters which contain the countries with higher GPDs. Let’s see if this relationship is significant within each cluster :

**Correlation for cluster 1

Between avg_gdp and women diabetes	0.368
Between avg_gdp and men diabetes	0.297

	Diabetes vs Women diabetes (cluster 1)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-1949.02	2590.14	-0.75	0.471
prop women diabetes	73395.77	61776.03	1.19	0.265
Observations	11
R2 / R2 adjusted	0.136 / 0.040

	Diabetes vs Men diabetes (cluster 1)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-1823.00	3150.00	-0.58	0.577
prop men diabetes	48218.14	51590.68	0.93	0.374
Observations	11
R2 / R2 adjusted	0.088 / -0.013

**Correlation for cluster 2

Between avg_gdp and women diabetes	-0.235
Between avg_gdp and men diabetes	0.325

	Diabetes vs Women diabetes (cluster 2)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	1113.78	1048.47	1.06	0.306
prop women diabetes	-15283.41	16887.89	-0.90	0.381
Observations	16
R2 / R2 adjusted	0.055 / -0.012

	Diabetes vs Men diabetes (cluster 2)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-901.34	834.85	-1.08	0.299
prop men diabetes	14396.62	11208.38	1.28	0.220
Observations	16
R2 / R2 adjusted	0.105 / 0.042

**Correlation for cluster 3

Between avg_gdp and women diabetes	-1
Between avg_gdp and men diabetes	-1

	Diabetes vs Women diabetes (cluster 3)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	1920.92	NaN	NaN	NaN
prop women diabetes	-47597.62	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

	Diabetes vs Men diabetes (cluster 3)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	1698.19	NaN	NaN	NaN
prop men diabetes	-24247.81	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

To go further, we decided to represent on a cluster map to see if there was a difference between the north and the south for example.

#> Simple feature collection with 37 features and 11 fields
#> Geometry type: MULTIPOLYGON
#> Dimension:     XY
#> Bounding box:  xmin: -63.1 ymin: -21.4 xmax: 55.8 ymax: 71.2
#> Geodetic CRS:  WGS 84
#> First 10 features:
#>    id LEVL_CODE NUTS_ID CNTR_CODE       NAME_LATN       NUTS_NAME
#> 1  AL         0      AL        AL       SHQIPËRIA       SHQIPËRIA
#> 2  AT         0      AT        AT      ÖSTERREICH      ÖSTERREICH
#> 3  BE         0      BE        BE BELGIQUE-BELGIË BELGIQUE-BELGIË
#> 4  NL         0      NL        NL       NEDERLAND       NEDERLAND
#> 5  PL         0      PL        PL          POLSKA          POLSKA
#> 6  PT         0      PT        PT        PORTUGAL        PORTUGAL
#> 7  DK         0      DK        DK         DANMARK         DANMARK
#> 8  DE         0      DE        DE     DEUTSCHLAND     DEUTSCHLAND
#> 9  EL         0      EL        EL          ELLADA          ΕΛΛΑΔΑ
#> 10 ES         0      ES        ES          ESPAÑA          ESPAÑA
#>    MOUNT_TYPE URBN_TYPE COAST_TYPE FID
#> 1           0         0          0  AL
#> 2           0         0          0  AT
#> 3           0         0          0  BE
#> 4           0         0          0  NL
#> 5           0         0          0  PL
#> 6           0         0          0  PT
#> 7           0         0          0  DK
#> 8           0         0          0  DE
#> 9           0         0          0  EL
#> 10          0         0          0  ES
#>                          geometry geo
#> 1  MULTIPOLYGON (((19.8 42.5, ...  AL
#> 2  MULTIPOLYGON (((15.5 48.9, ...  AT
#> 3  MULTIPOLYGON (((5.1 51.4, 5...  BE
#> 4  MULTIPOLYGON (((6.87 53.4, ...  NL
#> 5  MULTIPOLYGON (((19 54.4, 19...  PL
#> 6  MULTIPOLYGON (((-8.17 41.8,...  PT
#> 7  MULTIPOLYGON (((14.8 55.3, ...  DK
#> 8  MULTIPOLYGON (((8.64 54.9, ...  DE
#> 9  MULTIPOLYGON (((29.6 36.1, ...  EL
#> 10 MULTIPOLYGON (((4.29 39.8, ...  ES

The first cluster is therefore located more in the centre and north of Europe.

The second cluster would be located more to the southwest of Europe.

And the third cluster represents only two countries so it is hard to give it a region.

Clustering here doesn’t necessarily help us answer the research question. We can confirm with the previous statistical test without outliers that overall, the GDP of a country and its diabetes rate are negatively correlated.

However, the clusters defined above could help us answer our other research questions.

2.Do European countries that have higher GDPs consume less calories ?

As mentioned in the first point, countries with a higher GDP tend to have a lower diabetes rate, which could potentially be explained by the consumption of fewer calories.

But is there a real correlation between these two variables ? Let’s check :

**Correlation

Between avg_gdp and total calories

0.355

	avg_gdp vs total consumption
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-3017.67	1806.15	-1.67	0.106
total consumption	1.07	0.55	1.97	0.059
Observations	29
R2 / R2 adjusted	0.126 / 0.093

Neither the correlation between these two variables nor the linear regression is significant. However, it would be interesting to look further. First, we can see if using the GDP per person instead of the total average makes a difference.

The plot looks more or less the same as the one in the EDA section without the outliers. But is the relationship with this new variable statistically significant ?

**Correlation

Between avg_gdp per person and total calories

0.458

	avg_gdp per person vs total consumption
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-80976.25	41007.78	-1.97	0.059
total consumption	33.19	12.39	2.68	0.012
Observations	29
R2 / R2 adjusted	0.210 / 0.181

We see that the correlation with the variable gdp_per_personis higher now and the significance of the parameter in the regression, even though not high enough, increased.

Next we can proceed with an analysis within clusters, with the ones defined in the first question.

**Correlation for cluster 1

Between avg_gdp per person and total calories

-0.469

	avg_gdp per person vs total consumption (cluster 1)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	84886.45	27655.84	3.07	0.013
total consumption	-12.72	7.99	-1.59	0.146
Observations	11
R2 / R2 adjusted	0.220 / 0.133

**Correlation for cluster 2

Between avg_gdp per person and total calories

0.224

	avg_gdp per person vs total consumption (cluster 2)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-1941.55	18811.00	-0.10	0.919
total consumption	5.08	5.91	0.86	0.404
Observations	16
R2 / R2 adjusted	0.050 / -0.018

**Correlation for cluster 3

Between avg_gdp per person and total calories

1

	avg_gdp per person vs total consumption (cluster 3)
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-939315.96	NaN	NaN	NaN
total consumption	292.84	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

The results indicate that even within the clusters, these variables are not exactly correlated and the total calories consumed per person in a country is not a significant indicator of this country’s wealth.

When we look at our cluster plot, in terms of total calorie intake it is also the 1st cluster, which contains the richest countries considering GDP per capita, that consumes the most calories. One might therefore think that calorie consumption is not the main reason why high-GDP countries have lower diabetes rates.

3.How do the macronutrients (animal protein/plant protein/carbohydrates/fat) consumed differ between richer and poorer countries ?

We observed during the EDA that richer countries seemed to consume more fat on average. Now we want to see if we can confirm this relationship, and find out if there is a correlation between the average GDP of a country and the calories consumed for other macronutrients too.

Since the units of the variables “calories consumed from animal protein/plant protein/carbs/fat” are on a per capita basis, we think that including the GDP per capita in this part of the analysis could improve results. Additionally, we thought that another way of answering this research question could be to include the proportions of the calories consumed from different macronutrients on the total calories and see if it affects our analysis.

With these new metrics introduced, we will plot these relationships again.

It is also important to plot the clusters again to observe differences in the proportions of calories consumed.

We will be referring to these graphs and this cluster plot in our analysis in the next sections.

GDP - Animal protein

Let’s start with by analysing the relationship between a country’s wealth and it’s consumption of animal protein. We saw at the EDA section that there didn’t seem to be a trend between these 2 variables, overall the calories consumed from animal protein is very stable regardless of the average GDP of a country. We can confirm this by checking the correlation and by trying to fit a linear regression.

**Correlation

Between avg_gdp and calories from animal protein

0.272

	avg_gdp vs calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-716.41	860.87	-0.83	0.413
cal prot animal	5.28	3.59	1.47	0.153
Observations	29
R2 / R2 adjusted	0.074 / 0.040

Even though the correlation is positive, it is not very high. By looking at the summary of the linear regression, we can also see that the parameter is not significant. But what if we took the GDP per capita instead of the average GDP of a country ? The graph above seems to show again that the consumption of this macronutrient is very stable regardless of the GDP per capita of a country.

**Correlation

Between gdp per person and calories from animal protein

0.637

	Between gdp per person and calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-41157.10	16464.75	-2.50	0.019
cal prot animal	295.34	68.69	4.30	<0.001
Observations	29
R2 / R2 adjusted	0.406 / 0.384

This time, the correlation is higher and the variable’s estimate as a parameter is very significant ! Maybe the correlation can improve by considering proportion of calories instead of the calorie count.

**Correlation

Between gdp per person and proportion of calories from animal protein

0.536

	Between gdp per person and proportion of calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-41451.02	21463.09	-1.93	0.064
proportion animal prot	981450.06	297643.37	3.30	0.003
Observations	29
R2 / R2 adjusted	0.287 / 0.261

Here, the correlation is slightly lower than with the calorie count and the parameter’s estimate is a bit less significant but still quite solid. Now let’s see if we can find this positive relationship and better explanations within our defined clusters.

From our cluster plot above, we see that high GDP countries (clusters 1 and 3) tend to consume more calories of animal protein than low GDP countries, this is also the case for the proportions of calories consumed from animal protein. This follows what we have found so far. Are the correlations and linear regression in line with these observations ?

**Correlation for cluster 1

Between gdp per person and proportion of calories from animal protein

0.566

	Between gdp per person and proportion of calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	17379.15	11526.77	1.51	0.166
proportion animal prot	308366.99	149728.86	2.06	0.070
Observations	11
R2 / R2 adjusted	0.320 / 0.245

**Correlation for cluster 2

Between gdp per person and proportion of calories from animal protein

0.676

	Between gdp per person and proportion of calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-13068.03	8010.86	-1.63	0.125
proportion animal prot	404664.90	117758.56	3.44	0.004
Observations	16
R2 / R2 adjusted	0.458 / 0.419

**Correlation for cluster 3

Between gdp per person and proportion of calories from animal protein

1

	Between gdp per person and proportion of calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-61499.76	NaN	NaN	NaN
proportion animal prot	1812514.02	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

For each cluster, the correlation is positive and but still not so strong, and the estimate of the parameter for cluster 2 is significant.

In general, though it seems positive, we cannot conclude that there is a strong link between the wealth of a country and its animal protein consumption.

GDP - Plant protein

Now let’s move on to the analysis of the relationship between a country’s wealth and it’s consumption of plant protein. We saw at the EDA section, again, that there didn’t seem to be a trend between these 2 variables. The calories consumed from plant protein is very stable as we can see from the horizontal line, regardless of the average GDP of a country. To confirm this, we will check the correlation.

**Correlation

Between avg_gdp and calories from plant protein

0.043

	Between avg_gdp and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	276.48	1135.65	0.24	0.809
cal prot plant	1.52	6.75	0.23	0.823
Observations	29
R2 / R2 adjusted	0.002 / -0.035

As we can see, the correlation is very low and when we try to fit a linear model over these 2 variables, the parameter estimation is not significant at all. Let’s see if taking the GDP per capita makes a difference. In the graph above, the consumption of this macronutrient is very stable too.

**Correlation

Between gdp per person and calories from plant protein

-0.357

	Between gdp per person and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	78466.08	25364.62	3.09	0.005
cal prot plant	-299.07	150.66	-1.99	0.057
Observations	29
R2 / R2 adjusted	0.127 / 0.095

The correlation improves when we consider the GDP per capita instead of the average. We also see that the relationship between the 2 variables is rather negative. However the parameter estimation is still not significant enough, even though it improved. Can we find a better explanation by using the proportion of calories rather than the calorie count of the plant protein consumed ?

#> [1] -0.683

	Estimate	Std. Error	t value	Pr(>|t|)
(Intercept)	135049	22065	6.12	<0.001	***
GDP_diabetes_cal$proportion_plant_prot	-2103072	432612	-4.86	<0.001	***

**Correlation

Between gdp per person and proportion of calories from plant protein

-0.683

	Between gdp per person and proportion of calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	135049.20	22065.03	6.12	<0.001
proportion plant prot	-2103072.30	432611.97	-4.86	<0.001
Observations	29
R2 / R2 adjusted	0.467 / 0.447

The correlation improved immensely ! Now we can confirm that the relationship is rather negative, even though the correlation is rather moderate than strong. But now our parameter estimation as significant as possible. How does this translate to our clusters ?

The cluster plot above indicates that in countries with higher GDPs like in cluster 1 and 3, the proportions of calories consumed from plant protein are lower than in countries with lower GDPs. Let’s see if we can confirm these observations and see if these relationships are strong enough within each cluster.

**Correlation for cluster 1

Between gdp per person and proportion of calories from plant protein

-0.351

	Between gdp per person and proportion of calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	61779.35	18585.68	3.32	0.009
proportion plant prot	-438723.62	389754.96	-1.13	0.289
Observations	11
R2 / R2 adjusted	0.123 / 0.026

**Correlation for cluster 2

Between gdp per person and proportion of calories from plant protein

-0.417

	Between gdp per person and proportion of calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	38383.00	14155.73	2.71	0.017
proportion plant prot	-448628.20	261104.64	-1.72	0.108
Observations	16
R2 / R2 adjusted	0.174 / 0.115

**Correlation for cluster 3

Between gdp per person and proportion of calories from plant protein

1

	Between gdp per person and proportion of calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-398204.08	NaN	NaN	NaN
proportion plant prot	11498699.32	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

We see that the correlations within clusters is weaker. And actually, even in countries with lower GDPs that belong to cluster 2, the relationship is rather negative. This indicates us that visual interpretation is not enough for analysis. We can maybe say that less wealthy countries that are more developed than those in the same cluster as them may consume more plant protein than more wealthy countries but they still consume less of this macronutrient than their cluster mates.

Therefore, we cannot conclude that there is a significant link between the wealth of a country and its plant protein consumption, although this relationship seems negative.

GDP - Carbohydrates

Next, we will be evaluating the relationship of a country’s GDP and its carbohydrate consumption. In the EDA graph, there isn’t really trend in general between these 2 variables. The calories consumed from carbs is more or less stable for countries with an average GDP of around 500 billion dollars and more. Let’s check the correlation.

**Correlation

Between avg_gdp and calories from carbs

-0.083

	Between avg_gdp and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	1212.64	1588.16	0.76	0.452
cal carbs	-0.39	0.91	-0.43	0.669
Observations	29
R2 / R2 adjusted	0.007 / -0.030

The correlation is negative and very low. This could maybe be explained by the fact that variance is very high at the left part of the graph where the average GDP is lower, up until 500 billion dollars. The parameter estimation is not significant either, this regression is not appropriate to explain this relationship.

But what if we compared the 2 metrics at the same unit level like in the previous 2 sections ? Our graph above indicates that there may be a slight decrease of the caloric intake of carbs per capita the more a country’s GDP per capita is higher. Let’s see if this is true.

**Correlation

Between gdp per person and calories from carbs

-0.105

	Between gdp per person and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	49289.50	37855.69	1.30	0.204
cal carbs	-11.94	21.72	-0.55	0.587
Observations	29
R2 / R2 adjusted	0.011 / -0.026

Indeed, the correlation is higher now but it is still not strong enough. Our parameter is once again not significant to explain this new dependent variable either. But what if we changed our independent variable ? Because we see in the second graph above that this negative slope is even more accentuated when we take into account the proportion of the carbs consumed instead of their calorie count.

**Correlation

Between gdp per person and proportion of calories from carbs

-0.575

	Between gdp per person and proportion of calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	163176.52	36974.37	4.41	<0.001
proportion carbs	-255576.19	69975.38	-3.65	0.001
Observations	29
R2 / R2 adjusted	0.331 / 0.306

The correlation between these 2 variables is way more important, we can even say that their correlation is moderately strong. And the parameter estimation when we fit a linear model is quite significant. Let’s observe these relationships now within each cluster. According to our cluster plot, countries of cluster 1 and 3 (high GDP) consume a smaller proportion of carbs then those of cluster 2 (low GDP).

**Correlation for cluster 1

Between gdp per person and proportion of calories from carbs

0.553

	Between gdp per person and proportion of calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-28376.28	34846.14	-0.81	0.436
proportion carbs	137918.44	69276.82	1.99	0.078
Observations	11
R2 / R2 adjusted	0.306 / 0.229

**Correlation for cluster 2

Between gdp per person and proportion of calories from carbs

-0.751

	Between gdp per person and proportion of calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	66623.14	12383.47	5.38	<0.001
proportion carbs	-95782.01	22540.38	-4.25	0.001
Observations	16
R2 / R2 adjusted	0.563 / 0.532

**Correlation for cluster 3

Between gdp per person and proportion of calories from carbs

1

	Between gdp per person and proportion of calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-678064.39	NaN	NaN	NaN
proportion carbs	1532052.98	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

Surprisingly, with these correlations we see that the previous logic is inversed within clusters. For richer countries like those in cluster 1 and 3 the link between GDP and carb consumption is positive within their respective clusters, and for less rich countries like in those in cluster 2, it is negative. The estimate for the model parameter for cluster 2 is even very significant.

But overall, we cannot conclude that the link between the wealth of a country and its carb consumption is negative, as it is not strong enough.

GDP - Fats

For the last part of our 3rd research question, let’s analyse the relationship of a country’s GDP and its fat consumption. Once again when we take a look at the graph at the EDA section, at the left side of the graph, there seems to be a slight increase. But the calories consumed from fat is more or less stable for countries with an average GDP of around 500 billion dollars and more, just like for the carbs. What does the correlation coefficient tell us ?

**Correlation

Between avg_gdp and calories from fat

0.503

	Between avg_gdp and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-1941.46	827.39	-2.35	0.027
cal fat	2.13	0.70	3.03	0.005
Observations	29
R2 / R2 adjusted	0.253 / 0.226

From the get-go, the correlation is moderately high and positive. The estimation of the model parameter is also quite significant. Does taking the GDP per capita change our results ? In the first graph above, the fat consumption seems to very slightly increase when a country’s GDP is higher. Let’s check.

**Correlation

Between gdp per person and calories from fat

0.636

	Between gdp per person and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-46079.39	17640.70	-2.61	0.015
cal fat	64.23	14.98	4.29	<0.001
Observations	29
R2 / R2 adjusted	0.405 / 0.383

Our hypothesis was right. Not only did the correlation between the 2 variables increase, but the significance of the parameter estimation also improved. And how does switching the calorie count with the proportion of fat consumed per capita affect the correlation with the GDP per capita ? On the second graph above, we see that, just like for carbs, the positive slope becomes a bit more accentuated.

**Correlation

Between gdp per person and proportion of calories from fat

0.544

	Between gdp per person and proportion of calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-56316.18	25410.86	-2.22	0.035
proportion fat	241607.12	71783.98	3.37	0.002
Observations	29
R2 / R2 adjusted	0.296 / 0.269

This time, the correlation and the significance of the parameter goes down a little bit, even though not dramatically. Let’s see once more what these relationships are like within each cluster. By looking at the cluster plot, we can hypothesize that wealthier countries of cluster 1 and 3 consume a bigger proportion of fat then those of cluster 2 that are less wealthy.

**Correlation for cluster 1

Between gdp per person and proportion of calories from fat

-0.695

	Between gdp per person and proportion of calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	101563.63	20955.14	4.85	0.001
proportion fat	-162307.53	56002.68	-2.90	0.018
Observations	11
R2 / R2 adjusted	0.483 / 0.425

**Correlation for cluster 2

Between gdp per person and proportion of calories from fat

0.659

	Between gdp per person and proportion of calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-13798.71	8617.08	-1.60	0.132
proportion fat	84479.61	25779.67	3.28	0.006
Observations	16
R2 / R2 adjusted	0.434 / 0.394

**Correlation for cluster 3

Between gdp per person and proportion of calories from fat

-1

	Between gdp per person and proportion of calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	378543.88	NaN	NaN	NaN
proportion fat	-774350.31	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

Just like for carbs, surprisingly, we see again that the previous logic is inversed within clusters by looking at these correlations. For high GDP countries like those in cluster 1 and 3 the link between GDP and fat consumption is negative within their respective clusters, and for low GDP countries like in those in cluster 2, it is positive. The estimate for the model parameter for cluster 2 is again quite significant. This suggests that among rich countries, the countries that are the most rich tend to consume less fat, and vice versa.

In general, even though it seems positive, we cannot confidently say that there is a definite link between the wealth of a country and its fat consumption.

4.How do these differences in macronutrients relate to the diabetes prevalence in these countries ? What is the typical diet that can be observed in richer states that relates to lower diabetes prevalence ?

What we observed in the EDA seemed to not make sense to us as we were expecting a positive relationship between the total calories/calories consumed from fat and diabetes prevalence in a country. Let’s see now concretely if there is any correlation between the calories consumed from different macronutrients and the diabetes rate.

**Correlation

Between diabetes rate for men and calories from animal protein	-0.246
Between diabetes rate for women and calories from animal protein	-0.515

**Correlation

Between diabetes rate for men and calories from plant protein	0.227
Between diabetes rate for women and calories from plant protein	0.387

**Correlation

Between diabetes rate for men and calories from carbs	-0.032
Between diabetes rate for women and calories from carbs	0.155

**Correlation

Between diabetes rate for men and calories from fat	-0.404
Between diabetes rate for women and calories from fat	-0.697

But we also saw in the previous question that taking into account the proportions rather than the calorie count for the macronutrients gave us better results. Let’s see if the correlation with diabetes prevalence improves too.

**Correlation

Between diabetes rate for men and proportion of calories from animal protein	-0.141
Between diabetes rate for women and proportion of calories from animal protein	-0.399

**Correlation

Between diabetes rate for men and proportion of calories from plant protein	0.452
Between diabetes rate for women and proportion of calories from plant protein	0.706

**Correlation

Between diabetes rate for men and proportion of calories from carbs	0.272
Between diabetes rate for women and proportion of calories from carbs	0.616

**Correlation

Between diabetes rate for men and proportion of calories from fat	-0.303
Between diabetes rate for women and proportion of calories from fat	-0.621

It does ! So to answer this question, we will rather work with the variables related to the proportions of the different macronutrients consumed than the exact calories consumed.

We will now look in more detail at the relationship of each macronutrient with diabetes prevalence, starting with animal protein.

Diabetes - Animal protein

In the EDA, we observe a decreasing trend for the animal protein related to diabetes. The graph below with the proportions give us more or less the same result.

At the beginning of this question we saw that the correlation between these 2 variables is negative, but not very strong. This can be confirmed with the results of the linear regression below too.

	Between avg_gdp and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.08	0.01	6.19	<0.001
proportion animal prot	-0.13	0.17	-0.74	0.466
Observations	29
R2 / R2 adjusted	0.020 / -0.016

	Between avg_gdp and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.08	0.01	5.89	<0.001
proportion animal prot	-0.45	0.20	-2.26	0.032
Observations	29
R2 / R2 adjusted	0.159 / 0.128

So there is a negative trend for both men and women, but we also observe see that the effect is stronger for women. The linear regression does not indicate any relation since the parameters are not significant. Now let’s observe the relationship between diabetes and animal protein within each cluster.

**Correlation for cluster 1

Between men diabetes rate and calories from animal protein	-0.023
Between women diabetes rate and calories from animal protein	-0.114

	Between men diabetes rate and calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.06	0.02	3.66	0.005
proportion animal prot	-0.02	0.22	-0.07	0.947
Observations	11
R2 / R2 adjusted	0.001 / -0.111

	Between women diabetes rate and calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.05	0.01	3.39	0.008
proportion animal prot	-0.06	0.18	-0.35	0.738
Observations	11
R2 / R2 adjusted	0.013 / -0.097

**Correlation for cluster 2

Between men diabetes rate and calories from animal protein	0.386
Between women diabetes rate and calories from animal protein	-0.067

	Between men diabetes rate and calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.06	0.01	5.27	<0.001
proportion animal prot	0.25	0.16	1.57	0.140
Observations	16
R2 / R2 adjusted	0.149 / 0.088

	Between women diabetes rate and calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.06	0.01	7.97	<0.001
proportion animal prot	-0.03	0.12	-0.25	0.804
Observations	16
R2 / R2 adjusted	0.005 / -0.067

**Correlation for cluster 3

Between men diabetes rate and calories from animal protein	1
Between women diabetes rate and calories from animal protein	1

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between men diabetes rate and calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.05	NaN	NaN	NaN
proportion animal prot	1.42	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between women diabetes rate and calories from animal protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.02	NaN	NaN	NaN
proportion animal prot	0.72	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

For the first and second cluster, we have a negative relationship which is in line with the cluster graph, even though the correlation is not strong. It is also important to take into account that the regressions are not siginificant and it is therefore hard to draw anything from them.

Regarding the last cluster, we should expect a negative relationship but it seems to be positive. However, this correlation is very weak.

Diabetes - Plant protein

We saw at the EDA section, that there seemed to be a trend between these 2 variables. The calories consumed from plant protein seemed to increase the diabetes prevalence. As we can see in the graph below, it seems that with proportions, this slope is even more accentuated.

We can check if this relationship is significant with a linear regression.

	Between avg_gdp and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.03	0.01	2.49	0.019
proportion plant prot	0.69	0.26	2.63	0.014
Observations	29
R2 / R2 adjusted	0.204 / 0.175

	Between avg_gdp and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.02	0.01	-1.15	0.260
proportion plant prot	1.33	0.26	5.17	<0.001
Observations	29
R2 / R2 adjusted	0.498 / 0.479

The parameter is very significant when trying to explain the diabetes rate in women. With correlations at 0.452 and 0.706 respectively, these relationships are also moderately strong. Is it also the case for the clusters ?

**Correlation for cluster 1

Between men diabetes rate and calories from plant protein	0.535
Between women diabetes rate and calories from plant protein	0.690

	Between men diabetes rate and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.02	0.02	1.10	0.299
proportion plant prot	0.81	0.42	1.90	0.090
Observations	11
R2 / R2 adjusted	0.287 / 0.207

	Between women diabetes rate and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.00	0.01	0.10	0.923
proportion plant prot	0.85	0.30	2.86	0.019
Observations	11
R2 / R2 adjusted	0.476 / 0.418

**Correlation for cluster 2

Between men diabetes rate and calories from plant protein	-0.281
Between women diabetes rate and calories from plant protein	0.292

	Between men diabetes rate and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.09	0.02	5.67	<0.001
proportion plant prot	-0.33	0.30	-1.10	0.292
Observations	16
R2 / R2 adjusted	0.079 / 0.013

	Between women diabetes rate and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.05	0.01	4.48	0.001
proportion plant prot	0.23	0.20	1.14	0.272
Observations	16
R2 / R2 adjusted	0.085 / 0.020

**Correlation for cluster 3

Between men diabetes rate and calories from plant protein	1
Between women diabetes rate and calories from plant protein	1

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between men diabetes rate and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.31	NaN	NaN	NaN
proportion plant prot	9.02	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between women diabetes rate and calories from plant protein
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.15	NaN	NaN	NaN
proportion plant prot	4.60	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

For all of our clusters, according to the cluster graphs, we were expecting to see a positive relationship and our results mostly corresponded to those expectations. The correlations are quite strong for men and even higher for women in high GDP countries.

However in the second cluster, there is a negative correlation between the proportion of plant protein consumed and the diabetes prevalence in men, though not very strong.

Diabetes - Carbohydrates

In the EDA, it seems that carbohydrates have a rather flat trend compared to diabetes. Our graph with proportions below however, shows us a positive slope.

Indeed, the correlations between the proportion of carbs consumed and diabetes prevalence in men and women were positive, at 0.272 and 0.616 respectively. Let’s see the results when we try to fit a linear model.

	Between diabetes rate for men and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.04	0.02	1.70	0.101
proportion carbs	0.06	0.04	1.47	0.154
Observations	29
R2 / R2 adjusted	0.074 / 0.039

	Between diabetes rate for women and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.04	0.02	-1.65	0.110
proportion carbs	0.17	0.04	4.06	<0.001
Observations	29
R2 / R2 adjusted	0.380 / 0.357

We see that the linear regression for women is very significant, so increasing carbohydrates in a women’s diet would increase their diabetes rate. What do our clusters indicate ?

**Correlation for cluster 1

Between men diabetes rate and calories from carbs	-0.114
Between women diabetes rate and calories from carbs	0.125

	Between men diabetes rate and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.08	0.05	1.55	0.155
proportion carbs	-0.03	0.10	-0.34	0.738
Observations	11
R2 / R2 adjusted	0.013 / -0.097

	Between women diabetes rate and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.03	0.04	0.64	0.535
proportion carbs	0.03	0.08	0.38	0.715
Observations	11
R2 / R2 adjusted	0.016 / -0.094

**Correlation for cluster 2

Between men diabetes rate and calories from carbs	-0.309
Between women diabetes rate and calories from carbs	0.486

	Between men diabetes rate and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.10	0.02	5.05	<0.001
proportion carbs	-0.04	0.04	-1.21	0.245
Observations	16
R2 / R2 adjusted	0.095 / 0.031

	Between women diabetes rate and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.04	0.01	3.04	0.009
proportion carbs	0.05	0.02	2.08	0.057
Observations	16
R2 / R2 adjusted	0.236 / 0.181

**Correlation for cluster 3

Between men diabetes rate and calories from carbs	1
Between women diabetes rate and calories from carbs	1

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between men diabetes rate and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.53	NaN	NaN	NaN
proportion carbs	1.20	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between women diabetes rate and calories from carbs
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	-0.27	NaN	NaN	NaN
proportion carbs	0.61	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

For all of our clusters, we generally have positive correlations. This is what was expected by observing the cluster plot. Moreover, for the cluster 1 and 2 the correlations are negative in regards to diabetes prevalence in men. However these correlations are not strong at all, and as for our parameter estimations for the linear models we tried to fit over these variables, they are not significant.

Bur the positive trend between the consumption of carbs and diabetes prevalence is what makes the most sense to us so far, as foods that contain carbs like fast food etc tend to be bad for diabetes.

Diabetes - Fats

In the EDA part, the fats have a rather downward trend. Below, we have more or less the same graph with the proportions.

The correlation coefficients at the beginning of the question also showed us that there’s a negative relationship : -0.303 and -0.621 respectively for diabetes rates in men and women, in regards to the proportion of fat consumed. Let’s see what happens when we try to fit linear models.

	Between diabetes rate for men and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.09	0.01	6.40	<0.001
proportion fat	-0.07	0.04	-1.65	0.110
Observations	29
R2 / R2 adjusted	0.092 / 0.058

	Between diabetes rate for women and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.11	0.01	7.68	<0.001
proportion fat	-0.17	0.04	-4.12	<0.001
Observations	29
R2 / R2 adjusted	0.386 / 0.363

The parameter for fat is highly significant for women even though it only causes a little decrease to the diabetes rate. Are these results replicated in the clusters ?

**Correlation for cluster 1

Between men diabetes rate and calories from fat	0.016
Between women diabetes rate and calories from fat	-0.197

	Between men diabetes rate and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.06	0.04	1.68	0.128
proportion fat	0.00	0.09	0.05	0.962
Observations	11
R2 / R2 adjusted	0.000 / -0.111

	Between women diabetes rate and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.06	0.03	2.08	0.067
proportion fat	-0.05	0.08	-0.60	0.562
Observations	11
R2 / R2 adjusted	0.039 / -0.068

**Correlation for cluster 2

Between men diabetes rate and calories from fat	0.261
Between women diabetes rate and calories from fat	-0.508

	Between men diabetes rate and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.06	0.01	5.19	<0.001
proportion fat	0.04	0.04	1.01	0.329
Observations	16
R2 / R2 adjusted	0.068 / 0.001

	Between women diabetes rate and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.08	0.01	10.68	<0.001
proportion fat	-0.05	0.02	-2.21	0.044
Observations	16
R2 / R2 adjusted	0.258 / 0.205

**Correlation for cluster 3

Between men diabetes rate and calories from fat	-1
Between women diabetes rate and calories from fat	-1

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between men diabetes rate and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.30	NaN	NaN	NaN
proportion fat	-0.61	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

#> Warning: Model has zero degrees of freedom!

	Between women diabetes rate and calories from fat
Predictors	Estimates	std. Error	Statistic	p
(Intercept)	0.16	NaN	NaN	NaN
proportion fat	-0.31	NaN	NaN	NaN
Observations	2
R2 / R2 adjusted	1.000 / NaN

By observing the cluster plot, we can say that there’s a negative relationship between the proportion of fat consumed and diabetes prevalence within each cluster. In our results however that is not the case for the correlation with the diabetes rate in men for cluster 1 and 2, although they are weak. Otherwise, the results are in line with our expectations, but these are at most moderately strong relationships. Our regression results are also not very significant for every cluster.

Here, even though they are not sufficiently solid enough, we are surprised by our results because fatty foods are not good for diabetes, and our data shows that consuming more fat is not necessarily related to lower diabetes.

Typical Diet by Clusters

But then what does the typical diet of countries in each cluster look like ? Are these overall compositions what explain the differences in diabetes rates ?

We are going to investigate this very issue of calorie consumption patterns that could be related to the diabetes rate for each cluster. To see these patterns, we will take the average of every macronutrient in each cluster.

Mean for each cluster

	avg_gdp	gdp_per_person	prop_men_diabetes	prop_women_diabetes	cal_prot_animal	cal_prot_plant	cal_carbs	cal_fat	total_consumption
Cluster 1	1106	40938	0.061	0.042	262	164	1736	1292	3454
Cluster 2	167	14181	0.074	0.062	215	172	1738	1049	3174
Cluster 3	267	75791	0.059	0.035	263	143	1706	1355	3466

As we see from the cluster averages, cluster 2 has the highest rate of diabetes followed by cluster 1 and finally the cluster 3 with the lowest rate of diabetes.

There is not much difference between the diets of our different clusters. However, it should be noted that carbs represent 55% of the diet of cluster 2 and this food greatly influences the rate of diabetes, particularly among women. Indeed, there is a 1% increase in diabetes rate in women for every 0.168 carbs consumed on average per capita with a significant p-value at 1%. The proportion of carbs should therefore be reduced to less than half of the total proportion to try to have a negative impact on the diabetes rate.

Strangely enough, the proportion of fat is also lower in cluster 2 (33%) which is counter intuitive as its average diabetes rates are higher. But this result is consistent with the negative correlations of fat with diabetes rate calculated above. It is therefore complicated to draw a conclusion regarding the proportion of fat to be consumed to reduce the diabetes rate. Our hypothesis is that even if cluster 1 consumes more fat, it is composed of rich countries that have more prevention measures and certainly better hospital infrastructure.

It looks like richer countries in Europe like those of clusters 1 and 3 tend to consume a bit more animal protein and fats, and less rich countries like those of cluster 2 tend to consume more carbs.

Conclusion

• Take home message
• Limitations
• Future work?

Even though we tried using multiple methods and metrics like the correlation coefficient, trying to fit linear models and cluster analysis to answer our research questions, we couldn’t exactly find a lot of meaningful relationships between our various variables.

Oddly enough, it seems from our results that a higher consumption of animal protein could be related to a lower rate of diabetes, which would be counter-intuitive to M. Adeva-Andany’s (2019) article “Dietary habits contribute to define the risk of type 2 diabetes in humans”. But then again, our correlations were really not significant.

All macronutrients except protein are highly significant for women, therefore we make the assumption that malnutrition has a greater impact on the rate of diabetes in women than in men.

High GDP countries consume more fat, but have less diabetes and our results suggest that consuming more fat doesnt equal higher diabetes –> good fats, like good cholestrol ? look more into this

Maybe the low diabetes rate in high GDP countries is not about diet but exercise and healthcare, and maybe the relationship is not linear. Also we consider only Europe where economies are not that different after all.

Maybe our cluster number should have been 4, but then we would have the risk of having even smaller samples and less observations which are not good for statistical accuracy